Problem drinking is a significant public health problem associated with an array of negative consequences, and undergraduate college students are particularly vulnerable for extreme alcohol use and abuse. Despite significant advances in alcohol surveillance and prevention research, effectively and efficiently assessing the quantity, frequency, patterns and context in which individuals consume alcohol remains a methodological challenge. To address this challenge, we propose to develop and evaluate a state-of-the-art approach to daily retrospective alcohol assessment using wireless handheld computers. Our interdisciplinary team of experts in computer-based-data-collection and college student alcohol assessment propose to develop the Wireless Handheld Assessment Tool for Alcohol (WHAT-A) and to evaluate it on two college campuses (South Carolina State University and the University of South Carolina) with 280 undergraduate college students (140 African American and 140 White males and females), from 18-23 year of age. The evaluation study will use a randomized, two-period, crossover design, with baseline and two follow-up assessments to compare the WHAT-A to paper-and-pencil administered Drinking Self Monitoring Logs (DSML) and Timeline Followback assessments (TLFB). The specific aims of this study are: (1) To measure the relative differences in self-reported quantity, frequency, and context of alcohol consumption among individuals using WHAT-A, DSML, and TLFB; (2) To establish the reliability of alcohol assessment instruments administered using the WHAT-A; and, (3) To explore the potential of using the wireless handheld computers for health promotion interventions. Data analysis will include several types of repeated-measures analyses to compare the 3 assessment tools including the two-stage method for the two-period crossover. All repeated-measures analyses will be performed with a means model using SAS Proc Mixed software, providing separate estimates of the means for each assessment tool and for the mean difference between the methods of alcohol assessment. Additionally the daily count data on frequency and contexts of alcohol consumption (i.e., the 30-day within-subject drinking day events) will be analyzed using Poisson regression to help evaluate patterns of change in rate of alcohol consumption. Given the significant challenges surrounding high-risk drinking among college students and limitations of existing alcohol assessment instruments and approaches, there is a compelling need to explore the development of new innovative approaches for alcohol assessment. [unreadable] [unreadable]